The goal of this research is to elucidate the evolutionary dynamics and structural basis for the evolution of novel functions in a biomedically important gene family, the steroid hormone receptors. We will combine experimental evolution, crystallography, ancestral gene resurrection, and manipulative assays of gene function to analyze how receptors evolve tight molecular partnerships with novel hormonal ligands. The specific aims are: 1. Use an experimental evolution system in engineered yeast to select for human estrogen receptors which, like those in some mammary tumors, evolve to be stimulated by taxmoxifen, a major breast cancer drug. Using a battery of functional assays and evolutionary techniques for reconstructing the evolution of receptor sequences, we will determine the mechanistic basis for the evolution of this new ER-ligand interaction and the dynamics of receptor-ligand coevolution. 2. We will resurrect ancestral steroid receptors (using phylogenetic analysis and gene synthesis) and then evolve these receptors in this same experimental evolution system. Ancestral receptors will be selected to recapitulate the functional shifts in ligand specificity that occurred during real historical evolution over hundreds of millions of years. The mechanistic basis and evolutionary dynamics of this process will be studied in detail. 3. We will express, crystallize, and determine the three-dimensional structures of ancestral receptors, as well as those of several functionally divergent extant steroid receptors. These structures - together with the data on the evolution of new functions from Aims 1 and 2 - will provide a rich database from which to reconstruct the evolution of receptor protein structures. This is expected to reveal the structural mechanisms by which new receptor functions evolve, both in our experimental evolution system and in real historical evolution. This synthesis of manipulative, hypothesis-testing techniques will provide unprecedented detailed knowledge of the dynamics and mechanisms by which gene function evolves. These data should help resolve long-standing fundamental issues in evolutionary biology, and also improve our understanding of the structure-function relationships that determine steroid receptor function.